When the price elasticity data and predicted demand data aren\u2019t available, it creates a \u201cbounded price\u201d that allows for a range to be chosen to set a new price on the platform.<\/p>\n
Another patent recently granted to Walmart, US-12572954-B2, involves the use of machine learning to predict the demand of various items<\/a> and recommend prices. The schematic in the filing even shows that third-party data may be used to help determine the price, a controversial practice when it\u2019s employed by other businesses like airlines to set fares.<\/p>\n Walmart announced earlier this month that it had rolled out digital shelf labels that allow the company to quickly change the price consumers see in store, with the goal of every Walmart location utilizing the tech by the end of the year.<\/p>\n The patents awarded to Walmart involve e-commerce rather than brick and mortar locations, but it\u2019s easy to see how AI tools could be used in the future to create split-second changes to the price of goods based on any number of factors, like time of day or the number of people already in a given physical store.<\/p>\n Walmart didn\u2019t respond to questions from Gizmodo emailed on Thursday, but seemed to claim to the Financial Times<\/a> that the patents were \u201cunrelated to dynamic pricing.\u201d The logic, as best we can tell, is that one of the patents at issue \u201cwas specific to markdowns,\u201d perhaps trying to suggest that prices would only be lowered, though it\u2019s not clear.<\/p>\n The other claim was that another patent was \u201cdesigned for merchant teams to make decisions, not the technology,\u201d according to the Financial Times, which seems like a distinction without a difference when it comes to the practical reality of how this might be applied in the real world.<\/p>\n Hypothetically, it doesn\u2019t really matter if a store manager or someone in a central office is pinged to give final approval for a price adjustment. That simply looks like good quality control to benefit Walmart, not a serious protection for consumers from fast-paced price changes.<\/p>\n At least a dozen states are considering legislation to regulate dynamic or surveillance pricing this year, though New York is the only state that has passed a law on the topic. Consumers must be notified when an algorithm and personal data have been used to set a price, something that Washington Post subscribers recently received an email<\/a> about.<\/p>\n![\"Image](\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2026\/03\/US-12572954-B2.jpg\")
Image from Walmart\u2019s patent US-12572954-B2 showing how prices can be set with data about demand and third-party information. Image: USPTO <\/p>\n